How do I find standard deviations within bimodal data?

I have a bimodal data set that looks something like this:

From this, I have been using the function fitgmdist to find the mixing porportions and means of the two peaks:

    fit2 = fitgmdist(XM500,2)

    fit2 = 

    Gaussian mixture distribution with 2 components in 1 dimensions
    Component 1:
    Mixing proportion: 0.418376
    Mean:    0.4316

    Component 2:
    Mixing proportion: 0.581624
    Mean:    0.2965

I can pull the amplitude of each peak from the data presented, but I would be unable to re-create the two curves without a standard deviation. Does anyone have experience in this and know how it may be possible?

Thanks!

Answer 1

[Adam]

fit2.Sigma

will give you the covariance array which should just be variances if you have 1d data. The standard deviation can be calculated from this.

[Dan Crompton]

Thank you!